CN105548987A - Continuous wave radar object acceleration blind estimation method - Google Patents

Abstract

The invention provides a continuous wave radar object acceleration blind estimation method. The technical scheme comprises following three steps: a first step of sub-band resolution, a scattered sampling sequence of Doppler echo signals of the continuous wave radar in a random measuring period is divided into multiple sub-band signals; a second step of selecting an object echo sub-band signal, a sub-band signal with the largest quadratic sum is selected as the object echo sub-band signal; a third step of estimating the optimal acceleration, the acceleration is estimated by use of the object echo sub-band signal. The method has small calculated amount and strong real-time performance.

Description

A kind of continuous wave radar aimed acceleration blind estimating method

Technical field

The present invention relates to continuous wave radar signal process field, devise a kind of real-time target acceleration blind estimating method being applicable to continuous wave radar.

Background technology

The common continuous wave radar being widely used in target range measurement is a kind of single beam Narrow-band Radar.It can be used for the exterior trajectory measurement of the armament systems such as conventional weapon, rocket, guided missile, and can measure speed, the angle and distance of moving target, measuring accuracy is high simultaneously.

Usually, continuous wave radar, by certain hour interval (being called measuring period), intercepts one section of doppler echo data and makes corresponding data processing.Within shorter time period, the motion of the relative radar radial direction of target can be approximated to be uniformly accelerated motion, and the target Doppler echoed signal of continuous wave radar is approximately linear FM signal.By FFT (FastFourierTransformation, Fast Fourier Transform (FFT)) frequency spectrum analysis method such as algorithm is when analyzing doppler echo signal, if within the analysis of spectrum time period, the displacement that aimed acceleration causes is greater than 1/2nd of radar radio frequency wavelength, then before analysis doppler echo signal, need estimated acceleration and do corresponding compensation.

Existing conventional acceleration blind estimating method has: (1) record fly, Lou Shengqiang, hard, king's exhibition is at article " continuous wave radar acceleration and speed estimation method research [J] " (modern electronic technology, 2011,34 (21): 19 ~ 22) utilize cross-correlation technique estimated acceleration in, this method needs doppler echo signal to have higher signal to noise ratio (S/N ratio); (2) based on the method for acceleration template matches, aimed acceleration scope is split at regular intervals, each interval generates a compensation of phase sequence, analysis of spectrum is remake after phase compensation being done to doppler echo signal with it, this method estimated accuracy is high, but calculated amount is too large, is not suitable for measuring application in real time.Jiang Zhihong, Cao Yanwei, journey are flied, Huangfu may at article " the multi-frequency radar real time acceleration backoff algorithm [J] based on partial dechirping " (signal transacting, 2006,22 (5): 663 ~ 667) reduce calculated amount by extracting in, but the method needs the priori of target velocity.

Summary of the invention

Aimed acceleration blind estimating method provided by the invention is without the need to priori, counting yield is high, be applicable to the acceleration estimation in real-time measuring process target acquistion stage, or the quick detection of target maneuver characteristic in tracing process, be applicable to the use of the lower application scenario of signal to noise ratio (S/N ratio) simultaneously.

Technical scheme of the present invention is, a kind of continuous wave radar aimed acceleration blind estimating method, comprising:

The first step, sub-band division

If d (n), n=0,1,, N-1 is the discrete sampling sequence of doppler echo signal in continuous wave radar any one measuring period, d (n) is on average divided into M section, every section comprises B sampled point, and wherein N and B is the power side of 2, and the value of B is determined according to actual conditions.Then, m segment data is expressed as:

x(m,i)＝d(mB+i)

Wherein: m=0,1 ..., M-1, i=0,1 ..., B-1

For m=0,1 ..., M-1, utilizes following calculating discrete Fourier transformation:

$X(m,k)=\underset{i=0}{\overset{B-1}{\Σ}}x(m,i)e^{-j\frac{2\mathrm{\π}}{B}ik}$

Wherein: m=0,1 ..., M-1, k=0,1 ..., B-1

Such d (n) is broken down into the subband signal of B road M point, and wherein kth way band signal is:

{X(0,k),X(1,k),...,X(M-1,k)}

Following formula is utilized to calculate the quadratic sum P (k) of kth way band signal:

$P\left(k\right)=\underset{m=0}{\overset{M-1}{\Σ}}|X(m,k){|}^2,k=0,1,...,B-1$

Then obtain the quadratic sum of B subband signal altogether.

Second step, choose target echo subband signal

If: wherein: max{...} represents the maximal value choosing data in braces, t ₀certain integer in [0, B-1], and t ₀the quadratic sum of way band signal is maximum.Choose t ₀way band signal { X (m, t ₀), m=0,1 ..., M-1} is as target echo subband signal.

3rd step, estimate best acceleration

If known aimed acceleration interval is [A _down, A _up], to be averaged segmentation to this interval by interval D eltA, DeltA requires to determine according to estimated accuracy.For l=0,1 ..., L-1, wherein, [.] represents rounding operation, utilizes following formula sequence of calculation Y _l(r), and ask its maximal value A (l):

$Y_l\left(r\right)=\left|\underset{m=0}{\overset{M-1}{\mathrm{\Σ}}}X(m,t_0)e^{j\frac{2{\mathrm{\πa}}_l{B}^2{T}_s^2}{\mathrm{\λ}}{(m-\frac{M-1}{2})}^2}{e}^{-j2\mathrm{\π}\frac{rm}{M}}\right|,r=0,1,...,M-1$

A(l)＝max{Y _l(0),Y _l(1),...,Y _l(M-1)}

Wherein: a _l=A _down+ (l+0.5) × DeltA, T _sbe the sampling period of continuous wave radar doppler echo signal discrete sampling sequence d (n), λ is continuous wave radar radio frequency wavelength.

If: A _p0=max{A (0), A (1) ..., A (L-1) }, p ₀it is certain integer in [0, L-1].The aimed acceleration a then estimated _p0for: a _p0=A _down+ (p ₀+ 0.5) × DeltA.

The beneficial effect that the present invention obtains is: first the present invention makes sub-band division to N point discrete signal d (n), obtains B way band signal (first step see technical scheme).By the size select target echo subband signal of the quadratic sum of B way band signal, its number of data points be M (wherein second step see technical scheme), considerably reduce the calculated amount (the 3rd step see technical scheme) estimating best acceleration, make the method be applicable to being applied to the blind estimate of continuous wave radar target real time acceleration.Therefore, calculated amount of the present invention is little, real-time.

Accompanying drawing explanation

Fig. 1 is the process flow diagram of continuous wave radar aimed acceleration blind estimating method provided by the invention;

Fig. 2 is the comparison diagram of the acceleration that the acceleration that utilizes the present invention to obtain and speed difference method obtain;

Fig. 3 is the signal to noise ratio (S/N ratio) curve of acceleration compensation and the uncompensation doppler echo spectral line utilizing the present invention to obtain.

Embodiment

Process below in conjunction with accompanying drawing and measured data is further elaborated the present invention.

Fig. 1 is the process flow diagram of continuous wave radar aimed acceleration blind estimating method provided by the invention.

Fig. 2, Fig. 3 are the horizontal ordinates of measured data result figure, figure is all computation period sequence numbers.Correlation parameter is: measured data is certain rocket projectile X-band continuous wave radar echo data, and data time paragraph continues up to free flight phasel subsequently from initial boost section, sampling period T _s=4us.Data processing parameters: N=32768, M=128, B=256, DeltA=5m/s ².Fig. 2 is the comparison diagram of the acceleration that the acceleration that utilizes the present invention to obtain and speed difference method obtain, and ordinate unit is m/s ².The acceleration wherein utilizing the present invention to obtain calculates in real time in rocket projectile flight course, and the acceleration that speed difference method obtains utilizes target velocity curve to obtain afterwards.From figure, the acceleration of estimation and velocity contrast divide the acceleration obtained to be identical.Fig. 3 is the signal to noise ratio (S/N ratio) curve that the acceleration utilizing the present invention to obtain makes compensation and uncompensation doppler echo spectral line, ordinate unit is dB, in figure, solid line is through the signal to noise ratio (S/N ratio) curve of the doppler echo spectral line of acceleration compensation, and dotted line is not through the signal to noise ratio (S/N ratio) curve of the doppler echo spectral line of acceleration compensation.As can be seen from the figure, the signal to noise ratio (S/N ratio) through the doppler echo spectral line of acceleration compensation is improved, thus proves the accuracy of estimated acceleration of the present invention further.

Claims (1)

1. a continuous wave radar aimed acceleration blind estimating method, comprising:

The first step, sub-band division:

If d (n), n=0,1 ..., N-1, be the discrete sampling sequence of doppler echo signal in continuous wave radar any one measuring period, d (n) is on average divided into M section, and every section comprises B sampled point, and wherein N and B is the power side of 2; Then, m segment data is expressed as:

x(m,i)＝d(mB+i)

Wherein: m=0,1 ..., M-1, i=0,1 ..., B-1

Utilize following calculating discrete Fourier transformation:

$X(m,k)=\underset{i=0}{\overset{B-1}{\Σ}}x(m,i)e^{-j\frac{2\mathrm{\π}}{B}ik}$

Wherein: k=0,1 ..., B-1

Then kth way band signal is:

{X(0,k),X(1,k),...,X(M-1,k)}

Following formula is utilized to calculate the quadratic sum P (k) of kth way band signal:

$P\left(k\right)=\underset{m=0}{\overset{M-1}{\Σ}}|X(m,k){|}^2$

Second step, choose target echo subband signal:

If: wherein: max{...} represents the maximal value choosing data in braces, t ₀be certain integer in [0, B-1], choose t ₀way band signal { X (m, t ₀), m=0,1 ..., M-1} is as target echo subband signal;

3rd step, estimate best acceleration:

If known aimed acceleration interval is [A _down, A _up], by interval D eltA, this interval is averaged segmentation; For l=0,1 ..., L-1, wherein, [.] represents rounding operation, utilizes following formula sequence of calculation Y _l(r), and ask its maximal value A (l):

$Y_l\left(r\right)=\left|\underset{m=0}{\overset{M-1}{\Σ}}X\left(m,t_0\right)e^{j\frac{2{\mathrm{\πa}}_l{B}^2{T}_s^2}{\mathrm{\λ}}{\left(m-\frac{M-1}{2}\right)}^2}{e}^{-j2\mathrm{\π}\frac{rm}{M}}\right|,r=0,1,...,M-1$

A(l)＝max{Y _l(0),Y _l(1),...,Y _l(M-1)}

Wherein: a _l=A _down+ (l+0.5) × DeltA, T _sbe the sampling period of continuous wave radar doppler echo signal discrete sampling sequence d (n), λ is continuous wave radar radio frequency wavelength;

If A _p0=max{A (0), A (1) ..., A (L-1) }, p ₀certain integer in [0, L-1], then the aimed acceleration a estimated _p0=A _down+ (p ₀+ 0.5) × DeltA.